In aligning two parts employing machine vision, it is common to provide two spaced-apart holes in one part (called the "top-plate") and two targets on the second part (called the "substrate"). The two targets are spaced apart the same distance as the two holes.
The vision system looks at the substrate and determines the X,Y positions of the targets. The top-plate is then moved into the field of view and the vision system determines the X,Y position of the holes. The substrate and top-plate are then brought into alignment with each other (one or the other part being mounted on an alignment stage). Next, the substrate and top-plate are brought into assembly contact with each other by transforming the coordinates and determining the X, Y and theta motions required for the openings and the targets to coincide.
It is desired to have the vision system re-check the alignment. The reflectivities of the top-plate and the substrate are discernably different, as are the reflectivities of the substrate and the target. However, the reflectivities of the top-plate and the target are often similar. Consequently, upon bringing the top-plate and substrate together, the reflectivity of the target is similar enough to that of the top-plate that the vision system cannot easily locate the alignment holes. Thus, re-checking alignment is difficult at best.
Accordingly, a need remains for apparatus and a method which permits facile alignment of two parts by machine vision and convenient re-checking thereof.